Identifying and Forecasting Induced Seismicity Due to Wastewater Injection

Date: 4/26/2019

Time: 11:00 AM

Room: Cascade I

Wastewater disposal has been related to increases in seismicity in several parts of the Central United States. The vast majority of events have occurred in Oklahoma, which in 2015 experienced an almost 200-fold increase compared to the average annual seismicity rate before 2009, with the largest magnitudes reaching 5.8. In an attempt to quantify the extent of the problem, we develop a framework for seismic hazard assessment in regions with potentially induced seismicity. We developed a physics-based, semi-empirical model that simulates the observed seismicity given the injection time-history and can be used to distinguish the expected tectonic seismicity from disposal-induced seismicity with a certain degree of confidence. The model is also capable of forecasting seismicity rates in terms of space, time and magnitude distribution, given injection forecasts. Our proposed model is a modified version of the Gutenberg-Richter law, building upon the Seismogenic Index model that predicts a linear relationship between the number of induced events and the injected volume. We have allowed the injected volumes to diffuse in space and time and have introduced a time-lag that is dependent on injection rate and derived from rate-and-state friction concepts. The results show that most of the seismicity in Oklahoma is statistically identified as induced by wastewater injection (p-value < 0.1) and that the model is able to forecast both the peak and the decrease in seismicity, even when its calibration is limited to as early as late 2012.

Presenting Author: Iason Grigoratos

Authors